Complete guide to Google Search Console

Complete guide to Google Search Console, the free tool through which Google tells you how it sees your website: which searches display it, which pages are indexed and why others are not, what technical problems it has found, and whether there are manual actions. It is the only source of data from inside Google and, for us, the first thing opened in any audit.

This guide goes through each report in menu order, explaining what to look for and what decision to make. It assumes the property is already verified; if not, the steps are in the guide to connecting your website to Search Console.

Complete guide to Google Search Console

Property types

The domain property (domain) brings together all variants: http, https, www, subdomains. The URL prefix property covers only the exact variant. Use the domain property as your main source and, if you have language versions in subdirectories (/en/), additional URL prefix properties for each, to see the data separately. Domain property verification uses a TXT record in DNS, as we show in the guide to Zone Editor.

Performance report

Shows impressions, clicks, click-through rate (CTR) and average position over 16 months, with filters for query, page, country, device, search appearance and date. The data has a delay of two to three days and is sampled for rare queries (which is why the sum by query is smaller than the total). The filters we use in every analysis:

1. Compare 28 days with the previous 28 days, by page: what increased, what declined; declines on important pages are investigated first.
2. Queries containing the brand name versus those that do not: branded traffic tells you nothing about SEO; non-branded traffic does.
3. Positions between 4 and 15, with many impressions: pages "close to the first page", where improvement efforts have the greatest return.
4. CTR below the average for the position: titles and descriptions to rewrite; the method is described in the article on interaction signals.
5. Filter by page, then the Queries tab: all queries for a page; when a page receives impressions for two different topics, it is worth splitting it into two.
6. Mobile versus desktop: large differences in position indicate mobile experience problems.

The report also has separate tabs for Discover and Google News if the website receives traffic from them; Discover's particular features are covered in the article on optimising for Google Discover.

URL inspection

The search field at the top accepts any URL in the property and reports: whether it is indexed, when it was last crawled, which canonical Google chose, whether it was found through a sitemap or links, whether it is mobile-friendly, and what structured data was detected. The Test live URL button checks the page now, rather than from the index, which is useful after a fix. Request indexing requests a crawl, with a small daily quota; it is a request, not a guarantee. For any page that "does not appear in Google", start here.

Indexing

Page indexing report

Divides known URLs into indexed and not indexed, with the reason for each URL not indexed: Not found (404), Excluded by noindex, Duplicate without user-selected canonical, Crawled, currently not indexed, Discovered, currently not indexed, Blocked by robots.txt, Redirect, and the others. Each reason has a different fix; all are explained in the guide to interpreting coverage reports and, case by case, in the guide to common indexing errors. The Validate fix button starts a recheck after a fix, with a result in one to two weeks.

Video, sitemaps and removals

Video pages shows which videos have been indexed and why others have not. Sitemaps receives sitemaps and reports their status, as we describe in the guide to submitting the XML sitemap. Removals temporarily hides a URL from results (six months), for emergencies (information published by mistake); it does not replace actually removing the page.

Experience

Core Web Vitals groups URLs into good, needs improvement and poor, on mobile and desktop, with the metric responsible; HTTPS shows URLs served without encryption. Both use data from real users, so they appear only for websites with enough traffic. What each metric means and how to fix it are covered in the article on page experience signals.

Enhancements (structured data)

For each type of rich result detected on the website (breadcrumb, FAQ, product, article, organisation, review), a report appears with valid items, warnings and errors. Errors prevent rich results; warnings do not. Markup corrections are checked with the structured data validator, and Google's rules for each type are in the guide to schema markup.

Security and manual actions

Two reports to check on every visit, even though they are almost always empty: Manual actions (penalties applied by evaluators) and Security issues (compromised website, deceptive content, malware). Any entry here takes absolute priority; recovery steps are in the guide to reappearing in Google after a penalty.

Links

The Links report shows pages with the most external links, websites linking most, anchor texts, and internal links per page. It is not complete (Google shows a sample), but it is free and directly from the source. We use it for three things: to see which pages have attracted links naturally (and make more like them), to detect waves of spam links (anchors in other languages, large numbers of unknown websites), and to find important pages with few internal links, to fix within the website. The strategy is in the guide to link building strategies.

Settings

Crawl stats shows daily Googlebot requests, response time, returned codes and host status; it is the main report for the crawl budget. Users and permissions manages access (owner, full, restricted); give the agency or collaborator full access, not owner access, and revoke it at the end. Change of address notifies Google when the domain changes. Association links the property to Google Analytics and other services.

Exporting data

The interface shows at most 1.000 rows. For serious analysis, export to Google Sheets or connect the property to BigQuery (bulk export, free up to BigQuery's limits), which retains all data without sampling and without the 16-month limit. We use the Search Console API for clients' monthly reports, with comparisons by page group, the same ones we deliver as part of the SEO audit.

Five things Search Console does not tell you

• Exact real-time rankings: average position is an average across all impressions in the period, across all filtered countries and devices; a page ranked 3 in Romania and 40 in Germany has "position 20".
• Traffic from other search engines and AI: only Google, only Search, Discover and News. Traffic from ChatGPT, Perplexity or Bing is visible in Analytics, not here.
• Conversions: Search Console stops at the click. Linking with Google Analytics 4 through Association brings search data into behaviour reports, but with the same sampling.
• All queries: rare ones are hidden for privacy reasons; on small websites, the sum of clicks by query can be 30-50% below the total.
• Why a page declined: the report shows that it declined, not the cause. Look for the cause in algorithm updates, website changes and competition, following the method in the article on measuring the impact of an update.

Common interpretation errors

1. "Impressions increased, so it is good": impressions also increase when a page appears at position 60 for far more searches; track clicks and position for the queries that matter.
2. "Average position dropped after we published new content": new pages enter at poor positions and pull the average down, even though nothing has gone wrong; compare a fixed set of pages.
3. "We have 400 pages not indexed, it is a problem": many are normal (redirects, intentional noindex, duplicates with correct canonicals); only reasons indicating excluded valuable pages matter.
4. "Validate fix failed, so the fix does not work": validation fails if even one URL in the group still has the problem; check the remaining URLs, not the whole group.

The routine we recommend

1. Weekly: performance report, comparison with the previous week; Manual actions and Security issues.
2. At each publication: inspect the new URL, Request indexing if the page matters.
3. Monthly: Pages (reasons for non-indexing, trends), Core Web Vitals, Sitemaps, Enhancements.
4. Quarterly: Links, Crawl stats, review users with access.
5. After any technical change (theme, plugins, migration): Pages and Crawl stats over the following two weeks.

Automated emails from Search Console (increases in errors, indexing problems, manual actions) must reach someone who reads them; set the address in account preferences. Search Console tells you what happened; what you do with the information is the rest of the SEO work, which we describe in the step-by-step guide to optimising a website.

Sources

• Search Console Help, performance report: metrics, filters and data limits.
• Search Console Help, page indexing report: statuses and reasons for non-indexing.

Source references

• Ajutor Search Console, raportul de performanță
• Ajutor Search Console, raportul de indexare a paginilor